[openstack-dev] [cinder][nova] Are disk-intensive operations managed ... or not?

Duncan Thomas duncan.thomas at gmail.com
Thu Oct 23 11:15:31 UTC 2014


On 23 October 2014 08:30, Preston L. Bannister <preston at bannister.us> wrote:
> John,
>
> As a (new) OpenStack developer, I just discovered the "CINDER_SECURE_DELETE"
> option.
>
> As an *implicit* default, I entirely approve.  Production OpenStack
> installations should *absolutely* insure there is no information leakage
> from one instance to the next.
>
> As an *explicit* default, I am not so sure. Low-end storage requires you do
> this explicitly. High-end storage can insure information never leaks.
> Counting on high level storage can make the upper levels more efficient, can
> be a good thing.
>
> The debate about whether to wipe LV's pretty much massively depends on the
> intelligence of the underlying store. If the lower level storage never
> returns accidental information ... explicit zeroes are not needed.

The security requirements regarding wiping are totally and utterly
site dependent - some places care and are happy to pay the cost (some
even using an entirely pointless multi-write scrub out of historically
rooted paranoia) where as some don't care in the slightest. LVM thin
that John mentioned is no worse or better than most 'smart' arrays -
unless you happen to hit a bug, it won't return previous info.

That's a good default, if your site needs better then there are lots
of config options to go looking into for a whole variety of things,
and you should probably be doing your own security audits of the code
base and other deep analysis, as well as reading and contributing to
the security guide.



More information about the OpenStack-dev mailing list